The video covers examples of where various data structures are used in an ML context to highlight the importance of understanding algorithms and data structures in a data science role.
Category: Machine Learning
Avoiding Feedback Loops in Recommender Systems
Recommender systems often suffer from exposure bias, where we have customer feedback from only those items we actually recommend to the customers. Since only the shown items can collect positive feedback, we end up showing the same items again and again to the customers. This phenomemon is often called bias-amplification or a feedback loop. This…
How to do Kfold Crossvalidation for Temporal Data?
What is Temporal Leakage in ML Pipelines?
Feedback Loops: What causes Bias Amplification in Recommender Systems?
Gaussian Processes for Bayesian Hyperparameter Tuning
Click here for the python code to get started
Stratified Sampling for Imbalanced Datasets
Successive Halving For Grid Search
How to tune hyperparameters with Randomized Grid Search?
Randomized Grid Search is a variation of Grid Search that samples each parameter from a distribution. Conventional grid search evaluates the model at fixed combinations of parameter values and could be very inefficient as the complexity grows exponentially with the number of parameters. Randomized Grid search is a more efficient and an effective alternative.
How to find the Optimal Threshold from ROC curve?
This brief video talks about how the ROC curve is constructed and how one can find the optimal threshold for a classifier such as logistic regression, from the ROC curve.